Join our Newsletter — 33% off our NHI Course
Home› Guides› CyberArk Certificate Manager (Venafi) Configuration Guide: Policy, Discovery…
Guide Non-Human Identity (NHI)

CyberArk Certificate Manager (Venafi) Configuration Guide: Policy, Discovery and Machine Identity Automation

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 19 November 2026 15 min read
On this page

CyberArk Certificate Manager, the platform formerly sold as Venafi TLS Protect, is where many enterprises set the rules for every TLS certificate their machines hold, so its own configuration deserves the same discipline as a privileged access system. This guide explains how to structure issuing templates and applications, run discovery, lock down API keys and service accounts, use approvals and event logs, and prepare renewal automation for much shorter certificate lifetimes.

The product has been renamed more than once. CyberArk describes Certificate Manager, SaaS as formerly known as Venafi TLS Protect Cloud. Following the Palo Alto Networks acquisition of CyberArk, Palo Alto Networks states that CyberArk Certificate Manager is now Next-Gen Trust Security (NGTS), and that the self-hosted Certificate Manager, built on Trust Protection Foundation, continues alongside the self-hosted SSH Manager and Code Sign Manager products. This guide uses "Certificate Manager" for the SaaS service throughout, because that is the name most existing customers will find in their tenants and runbooks.

This is an independent guide from NHI Mgmt Group, based on CyberArk's public documentation as of November 2026. It is not sponsored or reviewed by CyberArk. Product features change, so confirm settings against CyberArk's current documentation.

Key takeaways

  • Issuing templates are the policy layer. CyberArk describes them as the way to lock down security policies for new certificates, with attributes pre-populated and locked, and separate templates per team and use case.
  • The platform holds powerful non-human identities of its own: API keys sent in the tppl-api-key header and service accounts. Treat each as a privileged credential with an owner, a scope and a rotation date.
  • Discovery, machine identities, auto-renewal monitoring and event logs are separate API resource groups. Configure all of them; a tenant that only issues certificates sees a fraction of the estate.
  • Certificate and revocation approvals exist as their own resource groups. Use them for high-risk templates rather than relying on self-service everywhere.
  • Shorter public TLS lifetimes make renewal automation mandatory. Build it now through the API, plugins and integrations, and measure it through the auto-renewal monitoring data.
  • The SaaS API is regional. Pin automation to the base URL for your tenant region and record it, so scripts do not silently fail or send credentials to the wrong endpoint.

What Certificate Manager manages and where trust sits

Certificate Manager is a control plane for machine identities: it does not replace your certificate authorities, it governs how certificates are requested from them, where they are installed and when they are renewed. The public API reference for the SaaS service lists the main building blocks. Under certificates and lifecycle it shows Certificates, Certificate Installations, Certificate Discovery, Certificate Import, Private Key Import, Certificate Request and Certificate Policy. Under applications and machines it lists Application, Machine Identities, Machine Types and Machines. Users and access cover Users, User Accounts, Teams, Service Accounts and Credential Management. Monitoring covers inventory monitoring, auto-renewal monitoring, expiration reports and event logs, and there are separate groups for certificate approvals, revocation approvals, webhooks, VSatellite, plugins and tags.

That list is a useful map of where trust concentrates. Three things matter most for security teams:

  • Policy objects (issuing templates and certificate policy) decide what a valid certificate looks like. Whoever can edit them can widen every future certificate.
  • Credentials to certificate authorities and to machines (held for issuance, installation and discovery) let the platform act on your infrastructure. Credential Management is a resource group in its own right for that reason.
  • Automation identities (API keys and service accounts) are used by pipelines, cert-manager style integrations and scripts. They usually outnumber human administrators.

The self-hosted product has a parallel structure. CyberArk's self-hosted documentation shows that SSH Manager for Machines can act as an SSH certificate authority, configured by a Trust Protection Foundation administrator, where issuance templates are described as a blueprint (policy) for certificate creation. The same design principle applies to both deployment models: policy is defined once, centrally, and requesters only choose from approved templates.

For the wider context of certificate lifecycle management, see the Machine Identity, PKI and Certificate Lifecycle Guide. For the open source Kubernetes controller that commonly requests certificates from Certificate Manager, see the cert-manager security guide, which this page does not repeat.

Designing issuing templates and applications

CyberArk's own description of issuing templates is short and precise: they let you lock down policy for new certificates, you can create different templates for different teams and use cases, and certificate attributes are pre-populated and locked so that certificate owners can request policy-compliant certificates directly through self-service. The configuration work is in deciding how many templates you need and what each one permits.

A template model that scales

NHI Mgmt Group recommends starting from use case rather than from team, because the same team often runs both an internet-facing service and an internal one with different trust requirements. A workable starting set is:

  • Public TLS: issued from a public CA, restricted to domains your organisation has validated, with the shortest validity your automation can support.
  • Internal TLS: issued from your private CA, restricted to internal domain suffixes, never wildcard by default.
  • Service mesh and workload: short-lived certificates for workload identity, issued only to automation, never to people.
  • Client authentication: for mutual TLS clients, with key usage limited to client authentication.
  • Exception: a tightly controlled template for legacy systems that cannot meet the standard, tied to an approval step and an expiry date for the exception itself.

For each template, record in your change system the certificate authority it uses, the allowed key types and sizes, the allowed domain patterns, the maximum validity and whether a private key may leave the requesting machine. Those are the settings that change risk, and the field names in the console change between releases, so keep the intent documented outside the product.

Applications as the unit of ownership

The Application resource group is how certificates are tied to the system and people that own them. Map each application to an owning team, attach only the templates that application needs, and make sure every certificate is linked to an application. Certificates with no application are the ones nobody renews and nobody revokes. Tags, which have their own API group, are useful for business attributes such as environment, data classification and cost centre, but ownership should sit in the application and team model rather than in free-text tags.

Changing a template safely

Template edits affect every future request. Treat them as production changes: peer review, a ticket reference, a test request in a non-production application first, and a check of the event log afterwards to confirm the change was recorded against the right administrator.

Discovery and inventory

Policy only governs certificates that pass through the platform. Discovery finds the rest. The API exposes Certificate Discovery and Certificate Import as separate groups, alongside Private Key Import and Certificate Inventory Monitoring, which reflects a common sequence: find certificates on the network or in cloud services, import them into inventory, assign them to applications, then bring them under managed renewal.

VSatellite appears as its own resource group in the API reference. Confirm in CyberArk's current documentation what your satellite deployment does and which credentials it holds. Treat any component you run inside your own network for discovery or installation as a privileged host: it sits inside your network, talks to a cloud control plane and may hold credentials for the systems it scans or installs onto. Patch it, restrict its outbound and inbound network paths to what the documentation requires, and monitor it like any other management server.

NHI Mgmt Group recommends running discovery in three passes:

  1. Breadth: scan all known address ranges and cloud accounts to count certificates and find the issuers you did not know about.
  2. Ownership: assign every discovered certificate to an application or to a holding application with a named triage owner and a deadline.
  3. Control: move owned certificates onto managed templates and renewal, and record the remainder as accepted exceptions with an end date.

Our guide How to build a TLS certificate inventory covers inventory fields and triage in more depth.

API keys, service accounts and teams

The Certificate Manager SaaS API authenticates requests with an API key sent in the tppl-api-key header, and the reference lists service accounts as an authentication option; HTTPS is required for all API requests. Every pipeline, Kubernetes issuer and integration that requests certificates therefore holds a credential that can obtain certificates under whatever templates its identity can use. That makes these credentials some of the most sensitive non-human identities in the estate.

Rules for automation identities

  • Prefer service accounts over personal API keys for any automation. A key tied to a named administrator breaks when that person leaves and carries that person's full permissions.
  • One identity per integration. A shared key across several pipelines makes it impossible to attribute a request or revoke one consumer.
  • Scope to applications and templates. A build pipeline for one service should only be able to request certificates for that service's application.
  • Store keys in a secrets manager, not in pipeline variables, repository files or container images, and inject them at run time.
  • Rotate on a fixed schedule and on staff change, and record the rotation date and owner in your NHI inventory.

Human roles and teams

Users, User Accounts and Teams are separate API groups. Use teams to express ownership and keep the number of tenant administrators small. Require single sign-on with phishing-resistant multi-factor authentication for administrators, review membership of administrative roles quarterly, and remove dormant accounts. The people who can edit templates, change certificate authority connections or create service accounts should be a different, smaller group from the people who request certificates.

Regional endpoints

The API reference lists separate base URLs by region, with https://api.venafi.cloud as the US default and regional hosts such as api.eu.venafi.cloud, api.au.venafi.cloud, api.uk.venafi.cloud, api.sg.venafi.cloud and api.ca.venafi.cloud. Hard-code the correct regional endpoint in each integration, allow only that host in egress rules for your build agents, and include it in your data residency records.

Approvals, revocation and event logs

Self-service is the main benefit of issuing templates, but not every template should be fully self-service. The API includes Certificate Approvals and Certificate Revocation Approvals as resource groups. NHI Mgmt Group recommends approval steps for:

  • wildcard certificates and certificates with many subject alternative names;
  • any template that issues from a public CA for a high-value domain;
  • the exception template described above;
  • revocation of certificates used by production customer-facing services, where an accidental revocation is an outage.

Approvals add friction, so keep them off routine renewals of already-approved certificates. Renewal of an existing, compliant certificate should be automatic.

The Event Logs group is the audit trail. Export event logs to your SIEM and alert on a short list of high-signal actions: template creation or change, certificate authority connection change, creation of a service account or API key, change to team or administrator membership, bulk revocation and bulk export of private keys where the template allows it. Webhooks, also in the API, can push events to downstream tools for faster response, but treat webhook endpoints as sensitive: they should authenticate the sender and should not be internet-exposed without controls.

Renewal automation for short certificate lifetimes

Public TLS certificate lifetimes are shrinking in stages, which turns renewal from an occasional task into a continuous one. The schedule and its operational effects are covered in our Certificate Expiry Outages and 47-Day TLS Certificates Guide. For Certificate Manager tenants, the practical question is which certificates still depend on a person.

The API groups give you the pieces to answer it. Certificate Installations and Machines describe where certificates are deployed; Certificate Auto-renewal Monitoring and Certificate Expiration Reports show what is and is not renewing on its own; Plugins extend issuance and installation to more targets. NHI Mgmt Group recommends:

  1. Report every certificate by renewal method: fully automated (request and install), request-only automated, or manual.
  2. Set a target for the share of public TLS certificates that are fully automated, and track it monthly from the auto-renewal monitoring data.
  3. Prioritise manual certificates on customer-facing services, then on internal services with long dependency chains.
  4. Reduce template validity in steps only once the automated share for that template is high enough that the next step will not create a wave of manual renewals.
  5. Rehearse emergency mass replacement, for example after a CA distrust event, using the same automation and measure how long it takes.

Kubernetes workloads commonly request certificates through cert-manager with a Certificate Manager issuer. Keep that issuer's credential scoped as described above, and follow the controller-side hardening in the cert-manager security guide. For rollout planning see the Certificate lifecycle manager rollout playbook.

Private keys, SSH certificates and the self-hosted product

The SaaS API includes Private Key Import, which means some templates and workflows let private keys pass through or be held by the platform. NHI Mgmt Group recommends a default of generating keys on the machine that uses them and sending only a certificate signing request, with central key generation allowed only for named legacy cases. Where keys are exported, record who exported them, where to and why, and alert on unusual export volumes.

Organisations using the self-hosted product for SSH should note how the SSH certificate authority is designed. The self-hosted documentation states that administrators can create multiple SSH CAs, each with multiple issuance templates, can generate certificates using an existing CA private key or a new one, can set access restrictions for clients and hosts, and can audit SSH certificate requests; the management interface does not itself issue certificates, which are issued through the WebSDK. Its example of a database administrator template that blocks root certificates while a system administrator template allows them is a good model: principals and extensions should be narrow per role. Our SSH Protocol and Certificate Authority Guide covers SSH CA design more broadly.

Protect CA private keys held by any self-hosted deployment with a hardware security module where your architecture supports it, and restrict WebSDK access to the automation identities that need it.

Where identity fits

Every certificate under management is a machine identity, and the platform itself is a privileged non-human system. Three identity questions should be answered in your design documents:

  • Who can change policy? Template and CA connection changes are equivalent to privileged access to every service that trusts those CAs. Govern them through your privileged access process.
  • Which machines can request which identities? API keys and service accounts must map one to one to integrations, with application scope.
  • Who owns each certificate? Every certificate should belong to an application and a team, so expiry, compromise and decommissioning have a responder.

CyberArk's wider platform hardening is covered separately in our CyberArk Security Hardening Guide.

Common failure patterns

  • One permissive template for everything. A single template that allows wildcards, long validity and any domain removes the point of central policy.
  • Personal API keys in pipelines. Automation breaks when the person leaves, and their full permissions sit in a build system.
  • Discovery without ownership. Thousands of discovered certificates sit in inventory with no application, so nobody acts on expiry warnings.
  • Unmonitored on-premises components. Satellite or self-hosted servers are patched less often than the systems they manage.
  • Event logs left in the console. Template changes and new service accounts are never reviewed because logs are not exported.
  • Shortening validity before automating. Lifetimes are cut while many certificates still renew manually, creating a recurring outage risk.
  • Wrong regional endpoint. Scripts are copied between regions and fail, or credentials are sent to an endpoint outside the intended data region.

Practitioner checklist

  • Confirm the current product name, tenant region and API base URL, and record them in your runbooks.
  • Define issuing templates by use case, with documented CA, key types, domain patterns and maximum validity for each.
  • Ensure every certificate belongs to an application with an owning team.
  • Run discovery across all networks and cloud accounts, and triage unowned certificates to a named owner.
  • Replace personal API keys in automation with per-integration service accounts scoped to applications.
  • Store all platform credentials in a secrets manager and rotate them on a schedule and on staff change.
  • Enforce SSO with phishing-resistant MFA for administrators and review administrative membership quarterly.
  • Enable approvals for wildcard, high-value public and exception templates, and for production revocations.
  • Export event logs to the SIEM and alert on template, CA connection, service account and administrator changes.
  • Patch and monitor any on-premises satellite or self-hosted servers as privileged hosts.
  • Report renewal method per certificate and track the share of fully automated public TLS certificates monthly.
  • Rehearse emergency mass certificate replacement and record the time taken.

Standards and references

Related NHI Mgmt Group resources: Machine Identity, PKI and Certificate Lifecycle Guide · cert-manager security guide · Certificate Expiry Outages and 47-Day TLS Certificates Guide · CyberArk Security Hardening Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 19 November 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.